Information attainable in some randomly incomplete data models |
| |
Institution: | 1. National Institute of Environmental Health Sciences, P.O. Box 12233, Research Triangle Park, Mail Drop A3-03, NC 27709, USA;2. Department of Biostatistics, and Statistics and Operations Research, 3101 McGavaran-Greenberg Hall, CB# 7420, Chapel Hill, NC 27599-7420, USA;1. McMaster University, Hamilton, Canada;2. University of Illinois at Chicago, USA;1. M.Sc. student of Computer Engineering, Imam Reza International University, Mashhad, Iran;2. Faculty of Computer Engineering and Information Technology, Sadjad University of Technology, Mashhad, Iran;1. University of Surrey, Guildford, Surrey, GU2 7XH, UK;2. William Harvey Research Institute, Barts and The London School of Medicine, Queen Mary University of London, London, EC1M6BQ, UK |
| |
Abstract: | The Fisher information is intricately linked to the asymptotic (first-order) optimality of maximum likelihood estimators for parametric complete-data models. When data are missing completely at random in a multivariate setup, it is shown that information in a single observation is well-defined and it plays the same role as in the complete-data model in characterizing the first-order asymptotic optimality properties of associated maximum likelihood estimators; computational aspects are also thoroughly appraised. As an illustration, the logistic regression model with incomplete binary responses and an incomplete categorical covariate is worked out. |
| |
Keywords: | |
本文献已被 ScienceDirect 等数据库收录! |
|